In this section we discuss a number of WSN applications that either fall in the C1WSN category or have a strong research–scientific focus. The applications discussed below are just a few examples of the many possible applications that exist or are evolving. The ability to deploy WSNs that interconnect in an effective manner with unattended WNs is expected to have a significant bearing on the efficacy of military and civil applications such as, but not limited to, combat field surveillance, security, and disaster management. These WSNs process data assembled from multi- ple sensors in order to monitor events in an area of interest. For example, in a disaster management event, a large number of sensors can be dropped by a helicopter; net- working these sensors can assist rescue operations by locating survivors, identifying risky areas, and making the rescue crew more aware of the overall situation and improving overall safety. Some WSNs have camera-enabled sensors; one can have aboveground full-color visible-light cameras as well as belowground infrared cameras. The use of WSNs will limit the need for military personnel involvement in dangerous reconnaissance missions. Security applications include intrusion detec- tion and criminal hunting [2.43]. Some examples of WSN applications are [2.6]:
1. Military sensor networks to detect and gain as much information as possible about enemy movements, explosions, and other phenomena of interest 2. Law enforcement and national security applications for inimical agent
tracking or nefarious substance monitoring (e.g., see Figure 2.9)
Figure 2.9 Law enforcement–national security application.
EXAMPLES OF CATEGORY 1 WSN APPLICATIONS 59
3. Sensor networks to detect and characterize chemical, biological, radiological, nuclear, and explosive (CBRNE) attacks and material
4. Sensor networks to detect and monitor environmental changes in plains, forests, oceans, and so on
5. Wireless traffic sensor networks to monitor vehicle traffic on highways or in congested parts of a city
6. Wireless surveillance sensor networks for providing security in shopping malls, parking garages, and other facilities
7. Wireless parking lot sensor networks to determine which spots are occupied and which are free
8. Borders monitoring with sensors and satellite uplinks
Figure 2.10 depicts the typical real-time administrative access to distributed WNs (motes) in an open-space sensor field. Real-time monitoring and sensor inter- rogation is typically supported. A number of illustrative examples are described in the subsections that follow. These examples just scratch the surface of the plethora of possible applications.
2.5.1 Sensor and Robots
Two technologies appear poised for a degree of convergence: mobile robotics and wireless sensor networks. Some researchers expect that mobile robotics will use WSNs to achieve ubiquitous computing environments. For example, Intel envisions mobile robots acting as gateways into wireless sensor networks, such as into the Smart Dust networks of wireless motes. These robots embody sensing, actuation, and basic (miniaturized) robotics functions. The field of mobile robotics deals with mechanical aspects (the wheels, motors, grasping arms, or physical layout) as well as with the logic aspects (the microprocessors, the software, and the telemetry). Two questions of interest are [2.15]:
Can a mobile robot act as a gateway into a wireless sensor network?
Can sensor networks take advantage of a robot’s mobility and intelligence?
Figure 2.10 Typical real-time administrative access to distributed motes.
To affect this convergence, inexpensive standards-based hardware, open-source operating systems, and off-the-shelf connectivity modules are required (e.g., Intel XScale microprocessors and Intel Centrino mobile technology).
One major issue with a mobile robot acting as a gateway is the communication between the robot and the sensor network. Some propose that a sensor network can be equipped with IEEE 802.11 capabilities to bridge the gap between robotics and wireless networks. For example, Intel recently demonstrated how a few motes equipped with 802.11 wireless capabilities can be added to a sensor network to act as wireless hubs [2.15]. Other motes in the network then utilize each other as links to reach the 802.11-equipped hubs; the hubs forward the data packets to the main 802.11-capable gateway, which is usually a PC or laptop. Using some motes as hubs reduces the number of hops that any one data packet has to make to reach the main gateway, and also reduces power consumption across the sensor network.
As an example, Intel recently installed small sensors in a vineyard in Oregon to monitor microclimates. The sensors measured temperature, humidity, and other factors to monitor the growing cycle of the grapes, then transmitted the data from sensor to sensor until the data reached a gateway. At the gateway, the data were interpreted and used to help prevent frostbite, mold, and other agricultural problems [2.15].
Intel, Carnegie Mellon University, University of Southern California, University of Pennsylvania, Northwestern, Georgia Tech, NASA, DARPA (the Defense Advanced Research Projects Agency), and NIST (National Institute of Standards and Technology) are just some of the institutions researching this topic. The Robotics Engineering Task Force (RETF; modeled after the Internet Engineering Task Force) has the goal of enabling government and university researchers to work collaboratively to establish standard software protocols and interfaces for robotics systems. The most pressing issue for the RETF is developing standards for commanding and controlling mobile robots.
Other examples of WSN applications include preventive maintenance for equipment in a semiconductor manufacturing fab, and sensor networks for theme parks. Both applications leverage the concept of heterogeneous WSNs, and both solve important business problems in their domains [2.15]. At7Intel’s semicon- ductor fabs, thousands of sensors track vibrations coming from various pieces of equipment to determine if the machines are about to fail. There is an established science that enables managers to determine the particular signature that a well- functioning machine should have. Typically, employees in the fab must gather the sensor data manually from each node—a costly and time-consuming process that is carried out periodically, on a schedule determined by the expected failure rate of the equipment. Going forward, networking the sensors could make the process more efficient and cost-effective. Intel reportedly plans to make use of the mote technology to build an application that acquires data automatically; pro- totypes have already been built (see Figure 2.11). Intel is also exploring the deployment of heterogeneous sensor networks in theme parks. Such networks
7The paragraphs that follow are based on Intel sources.
EXAMPLES OF CATEGORY 1 WSN APPLICATIONS 61
could be used for multiple purposes. One potential use is monitoring the quality of water in tanks (see Figure 2.12); currently, such monitoring is done manually; a WSN can make the process more accurate and efficient. Another potential use of the network is to provide Internet access to park visitors. Visitors can use the wireless network to reserve a space at a particular park attraction or to learn more about an exhibit. The wireless network could improve park management as well. Sensors could track attendance at park exhibits and rides, and manage- ment could use the network to access office applications from various stations throughout the park.
2.5.2 Reconfigurable Sensor Networks
Military applications require support for tactical and surveillance arrangements that employ reconfigurable sensor WNs that are capable of forming networks on the fly,
Fixed Vibration Sensors Motes
W I R E L E S S
W I R E L E S S Optional Repeater Motes
Gateway/
Collectors Ethernet
Ethernet Archive
Server
• Readings
• Filtering
• Windowing
• Data Analysis Digitized data are
dumped for analysis Analog data is
digitized Frequency:
0.5 Hz – 5 kHz
••
•
••
•
••
• PC
Figure 2.11 Intel fab environment with WSNs.
802.11 mesh network Sensor
Network
Internet
VPN Park Visitors Park Content Server
Sensor Data Servers
VPN-Corporate LAN Park Employees
Sensor Gateway
Figure 2.12 Theme park WSN example.
assembling themselves without central control, and being deployed incrementally.
Reconfigurable ‘‘smart’’ WNs are self-aware, self-configurable, and autonomous.
Self-organizing WSNs utilize mechanisms that allow newly deployed WNs to establish connectivity (to build up a network topology) spontaneously. Also, these networks have mechanisms for managing WN mobility (if any), WN recon- figuration, and WN failure (if and when that happens) [2.44].
2.5.3 Highway Monitoring
Transportation (traffic flow) is a sector that is expected to benefit from increased monitoring and surveillance. A specific example follows. (Traffic in the United States is growing at three times the rate of population growth and causing an esti- mated $75 billion lost annually due to traffic congestion.) Traffic Pulse Technology is an example of a WSN developed by Traffic.com [2.2,2.45]. The8goal of this system (which uses stationary WNs; see Figure 2.13) is to collect data through a sensor network, process and store the data in a data center, and distribute those data through a variety of applications. Traffic Pulse is targeted for open-air environments; it
8The rest of this subsection is based on [2.45].
Figure 2.13 Typical highway traffic-sensing installation. (Courtesy of Traffic.com.)
EXAMPLES OF CATEGORY 1 WSN APPLICATIONS 63
provides real-time collection of data (e.g., to check temperature or monitor pollution levels). The system is installed along major highways; the digital sensor network gathers lane-by-lane data on travel speeds, lane occupancy, and vehicle counts.
These basic data elements make it possible to calculate average speeds and travel times. The data are then transmitted to the data center for reformatting. The network monitors roadway conditions continuously on a 24/7 basis and provides updates to the data center in real time. The system collects key traffic information, including vehicle speeds, counts (volume), and roadway density, transmitting the data over a wireless network to a data center every 60 seconds.
In each major city, Traffic.com maintains a traffic pulse operations center that collects and reports on real-time event, construction, and incident data. This infor- mation supplements the data collected from the sensors. Each center produces the information through a wide range of methods: video, aircraft, mobile units, and monitoring of emergency and maintenance services frequencies. Applications include the following:
1. Private traffic information providers in the United States: The company’s real- time and archived data offer valuable tools for a variety of commercial and governmental applications.
2. Telematics: For mobile professionals and others, the company’s traffic information complements in-vehicle navigation devices, informing drivers not only how to get from point A to point B but how long it will take to get there, or even direct them to an alternative route.
2.5.4 Military Applications
A number of companies have developed WSNs that include customizable, sensor- laden, networked nodes and both mobile and Internet-hosted user interfaces [2.2,2.46]. For example, Rockwell Scientific’s wireless sensing network develop- ment system allows examination of issues relative to design, deployment, and use of microsensor networks. Wireless distributed microsensor networks consist of a collection of communicating nodes, where each node incorporates (1) one or more sensors for measuring the environment, (2) computing capability to process sensor data into ‘‘high-value’’ information and to accomplish local control, and (3) a radio to communicate information to and from neighboring nodes and eventually to external users. The company9has developed new prototype devel- opment platforms for experimenting with microsensor networks under a number of government- and industry-sponsored programs (see Figure 2.14). The baseline prototype wireless sensing unit is based on an open, modular design using widely available commercial-off-the-shelf (COTS) technology. These nodes combine sensors (such as mechanical vibration, acoustic, and magnetic) with a commercial digital cordless telephone radio and an embedded commercial RISC microprocessor in a small package.
9The rest of this subsection and Figure 2.14 are based on Rockwell Scientific sources [2.46].
Condition-Based Monitoring Again as an illustrative example, Rockwell Scien- tific is developing WSNs specifically tailored to the requirements for monitoring complex machinery and processes. Their WSNs have been deployed on board U.S. Navy ships as part of a developmental program with the Office of Naval Research. Exploratory studies have also been done for use of WSNs on aircraft, rotorcraft, and spacecraft as part of an overall integrated vehicle health manage- ment system. Machinery maintenance has evolved from run to fail (no mainte- nance) to scheduled maintenance (e.g., change oil every three months) to condition-based maintenance (CBM). All three techniques are in current use. The economic trade-off is between the cost of the CBM equipment and the staffing resources expended to determine the machine’s health and the cost of unexpected, as opposed to scheduled, repair and process downtime. With the emphasis of indus- try in the last couple of decades on just-in-time processes, unexpected machinery failure can be costly. The successful application of machinery monitoring programs can optimize the use of machinery and keep manufacturing costs in check by making the process more efficient [2.46]. The costs associated with CBM can be allocated into equipment, installation, and labor costs in collecting and analyz- ing the machine health data. WSNs are positioned to minimize all three costs and, in particular, to eliminate the staffing costs, which often are the largest. With the continuing advances in data processing hardware and RF transceiver hardware (cell phone markets drive this), the technology is now becoming available to install compact monitoring systems on machinery that avoid the installation expense of data cabling through RF link technology; these systems provide a mechanism for data acquisition and analysis on the monitoring unit itself. The primary challenge faced by WSNs for machinery and process monitoring is related to the quality of the information produced by both the individual sensors and the distributed sensor network. Nodes located on individual components must not only be able to provide information on the present state of the component (e.g., a bearing or gearbox), but also provide an indicator of the remaining useful life of the component [2.46].
The approach taken at Rockwell Scientific has been to mount two parallel efforts. Existing diagnostic routines and expert systems are being ported to WSN
Figure 2.14 Military examples. (Courtesy of Rockwell Scientific.)
EXAMPLES OF CATEGORY 1 WSN APPLICATIONS 65
hardware with modifications for autonomous data collection and analysis. The firm is also involved in developing advanced diagnostics algorithms for machinery vibration monitoring that provide advances over present systems. The main thrust in this area is to generalize diagnostic algorithms so that they do not depend on detailed knowledge of the machinery on which they are installed. Data-processing algorithms that determine critical machine parameters, such as the shaft speed or the number of rolling elements in a bearing, have been developed. The company is also developing the ability for distributed collections of WSN nodes located on machine components and/or throughout a process to provide information on the overall machine and/or process on which they are deployed. This is a primary advantage of a distributed sensing system in that it enables inferences from indivi- dual component data to be used to provide diagnostics for aspects of the system that are not being sensed directly. For example, monitoring bearing vibrations or motor currents can provide information not only on bearing health but also on the incep- tion and severity of pump cavitation. Pump cavitation, in turn, can provide informa- tion on the state of valves located throughout a pumping process.
The dynamically reconfigurable nature of WSNs is being exploited by Rockwell Scientific in an application of WSNs to space vehicle status monitoring in colla- boration with the Boeing Company. WSNs are deployed throughout space vehicles to perform a variety of missions during the different phases of the space flight. For example, during the launch phase, WSN nodes located on various critical compo- nents of the spacecraft can monitor vibration levels for out-of-compliance signals.
During flight and reentry, the WSN monitor structural disturbances caused by the significant temperature gradients encountered as different portions of the vehicle are alternately exposed and shadowed from the sun and atmosphere. This is accom- plished via coherent collection and processing of vibration and strain data. Upon landing, critical components will once again be monitored for out-of-compliance signals. These data are used to determine those components needing postflight maintenance or replacement, enabling faster turnaround for the space vehicle, thereby lowering costs [2.46].
Military Surveillance For military users, an application focus of WSN technol- ogy has been area and theater monitoring. WSNs can replace single high-cost sen- sor assets with large arrays of distributed sensors for both security and surveillance applications. The WSN nodes are smaller and more capable than sensor assets pre- sently in the inventory; the added feature of robust, self-organizing networking makes WSNs deployable by untrained troops in essentially any situation. Distrib- uted sensing has the additional advantages of being able to provide redundant and hence highly reliable information on threats as well as the ability to localize threats by both coherent and incoherent processing among the distributed sensor nodes.
WSNs can be used in traditional sensor network applications for large-area and perimeter monitoring and will ultimately enable every platoon, squad, and soldier to deploy WSNs to accomplish a number of mission and self-protection goals.
Rockwell Scientific has been working with the U.S. Marine Corps and U.S.
Army to test and refine WSN performance in desert, forest, and urban terrain.
For the urban terrain, WSNs are expected to improve troop safety as they clear and monitor intersections, buildings, and rooftops by providing continuous vigilance for unknown troop and vehicle activity. The primary challenge facing WSNs is accu- rate identification of the signal being sensed; one needs to develop state-of-the-art vibration, acoustic, and magnetic signal classification algorithms to accomplish this goal. Currently, WSNs run vibration detection algorithms based on energy thresh- olding; although this is a simple technique, it is subject to false alarms, leading to a desire for more sophisticated spectral signature algorithms. Low-power algorithms to classify a detected event as an impulsive event (e.g., either a footstep or gunshot) or vehicle (e.g., wheeled or tracked, light or heavy) have also been demonstrated.
The inclusion of multiple sensors on each node enables fusion of different sensed phenomenologies, leading to higher-quality information and decreased false alarm rates. Algorithms for fusing the seismic, acoustic, and magnetic sen- sors on a single node are being developed. Algorithms utilizing the advantages of a network of spatially separate nodes span a range of cooperative behaviors, each of which trades off detection quality versus energy consumption. Examples of cooperative fusion range from high-level decision corroboration (e.g., voting), to feature fusion, to full coherent beam formation. The examples discussed above are simply representative of many efforts under way at many companies involved in theater technology.
Borders Monitoring At press time Boeing Co. had secured a contract from the Department of Homeland Security to implement SBInet, the Secure Borders Initia- tive, along the northern and southern U.S. borders. The program was announced by DHS in 2005, and contracts were awarded in late 2006. The SBInet portion of the Secure Borders Initiative is the development of a technological infrastructure that facilitates the use of a variety of sensors and detection devices, and which enables that data to be forwarded to remote operations centers via Ku-band satellite uplinks.
2.5.5 Civil and Environmental Engineering Applications
Sensors can be used for civil engineering applications. Research has been under way in recent years to develop sensor technology that is applicable for buildings, bridges, and other structures. The goal is to develop ‘‘smart structures’’ that are able to self-diagnose potential problems and self-prioritize requisite repairs [2.47]. This tech- nology is attractive for earthquake-active zones. Although routine mild tremors may not cause visible damage, they can give rise to hidden cracks that could eventually fail during a higher-magnitude quake. Furthermore, after a mild earthquake, a buil- ding’s true structural condition may not be ostensively visible without some
‘‘below-the-skin’’ measurement. Smart Dust motes, tiny and inexpensive sensors devel- oped by UC–Berkeley engineers, are promising in this regard (see Figure 2.15). The battery-powered matchbox-sized WNs operating on TinyOS are designed to sense a number of factors, ranging from light and temperature (for energy-saving applications) to dynamic response (for civil engineering analysis) [2.47].
EXAMPLES OF CATEGORY 1 WSN APPLICATIONS 67